In the germ line of the nematode C. elegans, the epigenetic cycle of centromere inheritance is interrupted: the centromeric histone CENP-A is removed from chromatin at the mitosis-to-meiosis transition and must be re-established de novo during diplotene of meiosis I. A 2021 PLOS Biology study showed that the N-terminal tail of CENP-A is required specifically for this de novo establishment — through its interaction with the loading factor KNL-2 — while being dispensable for centromere maintenance, and that centromere identity, once set in one generation, is remembered for exactly one further generation through an unknown chromatin memory that reflects inherited information about centromere identity rather than maternally supplied full-length CENP-A protein. The questions the authors left open remain unresolved: the developmental timing and molecular trigger of identity determination, the mechanism by which the KNL-2/CENP-A N-terminal tail interaction creates a heritable chromatin state, the physical substrate of the one-generation 'centromere identity memory', and whether the chaperone LIN-53 participates in establishment in the endogenous germ line. The first three of these linked questions jointly constitute the mechanism of transgenerational centromere inheritance in this major model organism; the fourth is a related open question.

Background

Centromeres are the chromosomal regions on which kinetochores assemble; kinetochores are the multi-protein complexes that attach chromosomes to the spindle microtubules that pull them apart during cell division. In most eukaryotes, centromere identity is not specified by DNA sequence but inherited epigenetically: centromeric nucleosomes contain CENP-A (centromere protein A), a variant of histone H3 (the protein around which DNA is wrapped in nucleosomes), and pre-existing CENP-A nucleosomes serve as the mark that recruits new CENP-A after each DNA replication cycle, propagating centromere identity from one cell cycle to the next.

The nematode Caenorhabditis elegans is holocentric: instead of a single localized centromere, kinetochores form along the entire chromosome length, with CENP-A (called HCP-3 in this species) distributed over many dispersed loci. In the adult hermaphrodite germ line, the templated inheritance cycle is broken twice: CENP-A is removed from chromatin at the transition from mitotic germ cells to meiosis, and is re-loaded de novo during diplotene (a late sub-stage of prophase I of meiosis, when homologous chromosome pairs begin to separate but remain connected by recombination structures) in the proximal germ line; CENP-A localization additionally appears dynamic during the meiotic divisions after fertilization. CENP-A deposition in C. elegans depends on KNL-2, the ortholog of vertebrate M18BP1/MIS18BP1, a conserved SANTA- and Myb-domain protein that serves as the CENP-A loading factor.

A study by Prosée and colleagues (2021) used CRISPR/Cas9 knock-in worms lacking the N-terminal tail of CENP-A — the flexible region preceding the histone-fold domain — to dissect this process. The tail turns out to be required specifically for de novo establishment, not maintenance: homozygous tail-deletion mutants are viable in the first homozygous generation (F1) and segregate their chromosomes normally, because centromere identity established in the previous generation's germ line by full-length CENP-A is maintained for one generation even without the tail; but the F2 embryos die, with CENP-A and KNL-2 failing to adopt their normal opposition (bi-orientation) on mitotic chromosomes. This is a maternal-effect lethal profile (embryonic lethality determined by the molecular state of the mother rather than the zygote's own genotype), but the authors ruled out that F1 viability simply reflects maternally supplied full-length CENP-A protein, arguing instead that information about centromere identity — not the protein itself — is what is transferred from the maternal germ line into the zygote. Yeast two-hybrid and co-immunoprecipitation experiments mapped a direct interaction between the CENP-A N-terminal tail and the central region of KNL-2 (amino acids 269–470), independently confirmed by another group in the same year. LIN-53, the C. elegans homolog of the histone chaperone RbAp46/48 (a protein that escorts histones during assembly into chromatin), is known to maintain CENP-A chromatin association, and has separately been implicated in de novo CENP-A deposition on artificial chromosomes in embryos.

In other organisms the picture differs instructively: in mouse, centromere inheritance in early embryos passes through a 'plastic phase' governed by maternal CENP-A mRNA supply and repetitive centromere DNA rather than by the pre-existing CENP-A template, and in aging mouse oocytes existing CENP-A nucleosomes persist for months without new assembly. C. elegans therefore offers a genetically tractable case in which a chromatin state written in one generation is inherited into the next — analogous in kind to better-studied transgenerational marks such as Polycomb-mediated H3K27 methylation — but its timing, molecular mechanism, and memory substrate are unknown.

Problem Statement

Determine the mechanism by which centromere identity is established de novo and inherited from one generation to the next in the Caenorhabditis elegans maternal germ line. A complete answer must resolve the first three of the following linked aspects, which jointly constitute this mechanism; the fourth is a related open question raised by the source:

  1. Timing and trigger: the developmental stage and molecular trigger at which centromere identity is definitively established — whether within the diplotene/diakinesis window of meiosis I, when CENP-A is re-loaded onto chromatin, during the meiotic divisions after fertilization, when CENP-A localization is dynamic, or at another or overlapping stage.

  2. Establishment mechanism: how the interaction between KNL-2 and the CENP-A N-terminal tail converts into a heritable chromatin state that defines the centromere for an entire generation. Post-translational modifications of the tail and additional tail-interacting factors beyond KNL-2 are candidate components of such a mechanism but need not be exhaustively characterized.

  3. Memory substrate: the physical nature of the 'centromere identity memory' that allows centromere function to be maintained for one generation in the absence of the N-terminal tail — candidate substrates include a stable nucleosome arrangement, a specific chromatin modification, a higher-order or three-dimensional chromatin organization, a chaperone-associated state, or a combination of these — given that it is information rather than full-length CENP-A protein that is inherited.

  4. LIN-53's role (related open question): whether the histone chaperone LIN-53, known to be required for maintaining CENP-A chromatin association, also participates in establishment in the endogenous germ line.

These aspects are mutually constraining parts of one objective: the memory substrate of (3) is whatever state the mechanism of (2) writes during the developmental window pinned down by (1), with LIN-53 in (4) a candidate writer or maintainer. Available approaches in this system include C. elegans genetics (multi-generation crosses with tail-deletion and point-mutant alleles), timed perturbations (germline-staged depletion or rescue across the diplotene and post-fertilization windows), and interaction/mapping experiments (CENP-A deposition sites across generations).

References

  1. Prosée RF, Wenda JM, Özdemir I, Gabus C, Delaney K, Schwager F, Gotta M, Steiner FA (2021) Transgenerational inheritance of centromere identity requires the CENP-A N-terminal tail in the C. elegans maternal germ line. PLOS Biology 19(7):e3000968. doi:10.1371/journal.pbio.3000968, https://doi.org/10.1371/journal.pbio.3000968
  2. de Groot C, Houston J, Davis B, Gerson-Gurwitz A, Monen J, Lara-Gonzalez P, Oegema K, Shiau AK, Desai A (2021) The N-terminal tail of C. elegans CENP-A interacts with KNL-2 and is essential for centromeric chromatin assembly. Molecular Biology of the Cell 32(12):1193–1201. doi:10.1091/mbc.E20-12-0798, https://doi.org/10.1091/mbc.E20-12-0798
  3. Wenda JM, Prosée RF, Gabus C, Steiner FA (2021) Mitotic chromosome condensation requires phosphorylation of the centromeric protein KNL-2 in C. elegans. Journal of Cell Science 134(23):jcs259088. doi:10.1242/jcs.259088, https://doi.org/10.1242/jcs.259088
  4. Lin Z, Yuen KWY (2021) RbAp46/48 LIN-53 and HAT-1 are required for initial CENP-A HCP-3 deposition and de novo holocentromere formation on artificial chromosomes in Caenorhabditis elegans embryos. Nucleic Acids Research 49(16):9154–9173. doi:10.1093/nar/gkab217, https://doi.org/10.1093/nar/gkab217
  5. Wong CYY, Tsui HN, Wang Y, Yuen KWY (2024) Argonaute protein CSR-1 restricts localization of holocentromere protein HCP-3, the C. elegans CENP-A homolog. Journal of Cell Science 137(18):jcs261895. doi:10.1242/jcs.261895, https://doi.org/10.1242/jcs.261895
  6. Bellutti L, Macaisne N, El Mossadeq L, Ganeswaran T, Canman JC, Dumont J (2024) Regulation of outer kinetochore assembly during meiosis I and II by CENP-A and KNL-2/M18BP1 in C. elegans oocytes. Current Biology 34(21):4853–4868. doi:10.1016/j.cub.2024.09.004, https://doi.org/10.1016/j.cub.2024.09.004
  7. Wang Y, Yu Wong CY, Yuen KWY (2025) Aurora B/AIR-2 regulates sister centromere resolution and CENP-A/HCP-3 organization to prevent merotelic attachments. Journal of Molecular Cell Biology 16(10):mjae045. doi:10.1093/jmcb/mjae045, https://doi.org/10.1093/jmcb/mjae045
  8. Hollis JA, Stonick JA, Topalidou I, Young JM, Moens CB, Lehrbach NJ, Campbell MG, Malik HS (2026) Remote homology and functional genetics unmask deeply preserved Scm3/HJURP orthologs in metazoans. Science Advances 12(27):eaeg5356. doi:10.1126/sciadv.aeg5356, https://doi.org/10.1126/sciadv.aeg5356
  9. Özdemir I, Steiner FA (2022) Transmission of chromatin states across generations in C. elegans. Seminars in Cell & Developmental Biology 127:133–141. doi:10.1016/j.semcdb.2021.11.008, https://doi.org/10.1016/j.semcdb.2021.11.008
  10. Wang Y, Wu L, Yuen KWY (2022) The roles of transcription, chromatin organisation and chromosomal processes in holocentromere establishment and maintenance. Seminars in Cell & Developmental Biology 127:79–89. doi:10.1016/j.semcdb.2022.01.004, https://doi.org/10.1016/j.semcdb.2022.01.004
  11. Stirpe A, Heun P (2023) The ins and outs of CENP-A: Chromatin dynamics of the centromere-specific histone. Seminars in Cell & Developmental Biology 135:24–34. doi:10.1016/j.semcdb.2022.04.003, https://doi.org/10.1016/j.semcdb.2022.04.003
  12. Fukagawa T, Kakutani T (2023) Transgenerational epigenetic control of constitutive heterochromatin, transposons, and centromeres. Current Opinion in Genetics & Development 78:102021. doi:10.1016/j.gde.2023.102021, https://doi.org/10.1016/j.gde.2023.102021
  13. Das A, Iwata-Otsubo A, Destouni A, Dawicki-McKenna JM, Boese KG, Black BE, Lampson MA (2022) Epigenetic, genetic and maternal effects enable stable centromere inheritance. Nature Cell Biology 24(5):748–756. doi:10.1038/s41556-022-00897-w, https://doi.org/10.1038/s41556-022-00897-w
  14. Das A, Boese KG, Tachibana K, Baek SH, Lampson MA, Black BE (2023) Centromere-specifying nucleosomes persist in aging mouse oocytes in the absence of nascent assembly. Current Biology 33(17):3759–3765. doi:10.1016/j.cub.2023.07.032, https://doi.org/10.1016/j.cub.2023.07.032
  15. Caro L, Raman P, Steiner FA, Ailion M, Malik HS (2022) Recurrent but Short-Lived Duplications of Centromeric Proteins in Holocentric Caenorhabditis Species. Molecular Biology and Evolution 39(10):msac206. doi:10.1093/molbev/msac206, https://doi.org/10.1093/molbev/msac206